lib/wiki.php

   1 <?php
   2
   3 ///////////////////////////////////////////////////////////////////////////
   4 // wiki.php - class for Wiki style formatting
   5 //
   6 // Transforms input string with Wiki style formatting into HTML
   7 //
   8 //
   9 ///////////////////////////////////////////////////////////////////////////
  10 //                                                                       //
  11 // NOTICE OF COPYRIGHT                                                   //
  12 //                                                                       //
  13 //                                                                       //
  14 // Copyright (C) 2003 Howard Miller - GUIDE - University of Glasgow
  15 // guide.gla.ac.uk
  16 //                                                                       //
  17 // This program is free software; you can redistribute it and/or modify  //
  18 // it under the terms of the GNU General Public License as published by  //
  19 // the Free Software Foundation; either version 2 of the License, or     //
  20 // (at your option) any later version.                                   //
  21 //                                                                       //
  22 // This program is distributed in the hope that it will be useful,       //
  23 // but WITHOUT ANY WARRANTY; without even the implied warranty of        //
  24 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the         //
  25 // GNU General Public License for more details:                          //
  26 //                                                                       //
  27 //          http://www.gnu.org/copyleft/gpl.html                         //
  28 //                                                                       //
  29 ///////////////////////////////////////////////////////////////////////////
  30
  31 // set this to 1 *IF* we are running in Moodle
  32 // it enables Moodle specific functions, otherwise 0
  33 define( "IN_MOODLE",1 );
  34
  35 // state defines
  36 define( "STATE_NONE",1 ); // blank line has been detected, so looking for first line on next para
  37 define( "STATE_PARAGRAPH",2 ); // currently processing vanilla paragraph
  38 define( "STATE_BLOCKQUOTE",3 ); // currently processing blockquote section
  39 define( "STATE_PREFORM",4 ); // currently processing preformatted text
  40 define( "STATE_NOTIKI",5 ); // currently processing preformatted / no formatting
  41
  42 // list defines
  43 define( "LIST_NONE", 1 ); // no lists active
  44 define( "LIST_UNORDERED", 2 ); // unordered list active
  45 define( "LIST_ORDERED", 3 ); // ordered list active
  46 define( "LIST_DEFINITION", 4 ); // definition list active
  47
  48
  49 class Wiki {
  50
  51   var $block_state;
  52   var $list_state;
  53   var $list_depth;
  54   var $spelling_on;
  55   var $list_backtrack;
  56   var $output; // output buffer
  57
  58   function close_block( $state ) {
  59     // provide appropriate closure for block according to state
  60
  61     // if in list close this first
  62     $lclose = "";
  63     if ($this->list_state != LIST_NONE) {
  64       $lclose = $this->do_list( " ",true );
  65     }
  66
  67     $sclose = "";
  68     switch ($state) {
  69       case STATE_PARAGRAPH:
  70         $sclose =  "</p>\n";
  71         break;
  72       case STATE_BLOCKQUOTE:
  73         $sclose =  "</blockquote>\n";
  74         break;
  75       case STATE_PREFORM:
  76         $sclose =  "</pre>\n";
  77         break;
  78       case STATE_NOTIKI:
  79         $sclose =  "</pre>\n";
  80         break;
  81     }
  82
  83     return $lclose . $sclose;
  84   }
  85
  86
  87   function do_replace( $line, $mark, $tag ) {
  88     // do the regex thingy for things like bold, italic etc
  89     // $mark is the magic character, and $tag the HTML tag to insert
  90
  91     // BODGE: replace inline $mark characters in places where we want them ignored
  92     // they will be put back after main substitutue, stops problems with eg, and/or
  93     $bodge = chr(1);
  94     $line = eregi_replace( '([[:alnum:]])'.$mark.'([[:alnum:]])', '\\1'.$bodge.'\\2',$line );
  95
  96     $regex = '(^| |[(.,])'.$mark.'([^'.$mark.']*)'.$mark.'([^[:alnum:]]|$)';
  97     $replace = '\\1<'.$tag.'>\\2</'.$tag.'>\\3';
  98     $line = eregi_replace( $regex, $replace, $line );
  99
 100     // BODGE: back we go
 101     $line = eregi_replace( $bodge, $mark, $line );
 102
 103     return $line;
 104   }
 105
 106   function do_replace_sub( $line, $mark, $tag ) {
 107     // do regex for subscript and superscript (slightly different)
 108     // $mark is the magic character and $tag the HTML tag to insert
 109
 110     $regex = $mark.'([^'.$mark.']*)'.$mark;
 111     $replace = '<'.$tag.'>\\1</'.$tag.'>';
 112     return eregi_replace( $regex, $replace, $line );
 113   }
 114
 115   function do_list( $line, $blank=false ) {
 116     // handle line with list character on it
 117     // if blank line implies drop to level 0
 118
 119     // get magic character and then delete it from the line if not blank
 120     if ($blank) {
 121       $listchar="";
 122       $count = 0;
 123     }
 124     else {
 125       $listchar = $line{0};
 126       $count = strspn( $line, $listchar );
 127       $line = eregi_replace( "^[".$listchar."]+ ", "", $line );
 128     }
 129
 130     // find what sort of list this character represents
 131     $list_tag = "";
 132     $item_tag = "";
 133     $list_style = LIST_NONE;
 134     switch ($listchar) {
 135       case '*':
 136         $list_tag = "ul";
 137         $item_tag = "li";
 138         $list_style = LIST_UNORDERED;
 139         break;
 140       case '#':
 141         $list_tag = "ol";
 142         $item_tag = "li";
 143         $list_style = LIST_ORDERED;
 144         break;
 145       case ';':
 146         $list_tag = "dl";
 147         $item_tag = "dd";
 148         $list_style = LIST_DEFINITION;
 149         break;
 150       case ':':
 151         $list_tag = "dl";
 152         $item_tag = "dt";
 153         $list_style = LIST_DEFINITION;
 154         break;
 155       }
 156
 157     // tag opening/closing regime now - fun bit :-)
 158     $tags = "";
 159
 160     // if depth has reduced do number of closes to restore level
 161     for ($i=$this->list_depth; $i>$count; $i-- ) {
 162       $close_tag = array_pop( $this->list_backtrack );
 163       $tags = $tags . $close_tag;
 164       }
 165
 166     // if depth has increased do number of opens to balance
 167     for ($i=$this->list_depth; $i<$count; $i++ ) {
 168       array_push( $this->list_backtrack, "</$list_tag>" );
 169       $tags = $tags . "<$list_tag>";
 170     }
 171
 172     // ok, so list state is now same as style and depth same as count
 173     $this->list_state = $list_style;
 174     $this->list_depth = $count;
 175
 176     // apply formatting to remainder of line
 177     $line = $this->line_replace( $line );
 178
 179     if ($blank) {
 180       $newline = $tags;
 181     }
 182     else {
 183       $newline = $tags . "<$item_tag>" . $line . "</$item_tag>";
 184     }
 185
 186     return $newline;
 187   }
 188
 189   function line_replace( $line ) {
 190     // return line after various formatting replacements
 191     // have been made - order is vital to stop them interfering with each other
 192
 193     if (IN_MOODLE==1) {
 194       global $CFG;
 195     }
 196
 197     // ---- (at least) means a <HR>
 198     $line = eregi_replace( "^-{4}.*", "<div class=\"hr\"><hr /></div>", $line );
 199
 200     // is this a list line (starts with * # ; :)
 201     if (eregi( "^([*]+|[#]+|[;]+|[:]+) ", $line )) {
 202       $line = $this->do_list( $line );
 203     }
 204
 205    // typographic conventions
 206     $line = str_replace( "--", "&#8212;", $line );
 207     $line = str_replace( " - ", " &#8211; ", $line );
 208     $line = str_replace( "...", " &#8230; ", $line );
 209     $line = str_replace( "(R)", "&#174;", $line );
 210     $line = str_replace( "(r)", "&#174;", $line );
 211     $line = str_replace( "(TM)", "&#8482;", $line );
 212     $line = str_replace( "(tm)", "&#8482;", $line );
 213     $line = str_replace( "(C)", "&#169;", $line );
 214     // $line = str_replace( "(c)", "&#169;", $line );
 215     $line = str_replace( "1/4", "&#188;", $line );
 216     $line = str_replace( "1/2", "&#189;", $line );
 217     $line = str_replace( "3/4", "&#190;", $line );
 218
 219     $line = eregi_replace( "([[:digit:]]+[[:space:]]*)x([[:space:]]*[[:digit:]]+)", "\\1&#215;\\2", $line ); // (digits) x (digits) - multiply
 220
 221     // do formatting tags
 222     // NOTE: The / replacement  *has* to be first, or it will screw the
 223     // HTML tags that are added by the other ones
 224     $line = $this->do_replace( $line, "/", "em" );
 225     $line = $this->do_replace( $line, "\*", "strong" );
 226     $line = $this->do_replace( $line, "\+", "ins" );
 227     $line = $this->do_replace( $line, "-", "del" );
 228     $line = $this->do_replace_sub( $line, "~", "sub" );
 229     $line = $this->do_replace_sub( $line, "\^", "sup" );
 230     $line = $this->do_replace( $line, "\"", "q" );
 231     // $line = $this->do_replace( $line, "'", "q" );
 232     $line = $this->do_replace( $line, "%", "code" );
 233     $line = $this->do_replace( $line, "@", "cite" );
 234
 235     // convert urls into proper link with optional link text URL(text)
 236     $line = eregi_replace("([[:space:]]|^)([[:alnum:]]+)://([^[:space:]]*)([[:alnum:]#?/&=])\(([^)]+)\)",
 237       "\\1<A HREF=\"\\2://\\3\\4\" TARGET=\"newpage\">\\5</A>", $line);
 238     $line = eregi_replace("([[:space:]])www\.([^[:space:]]*)([[:alnum:]#?/&=])\(([^)]+)\)",
 239       "\\1<A HREF=\"http://www.\\2\\3\" TARGET=\"newpage\">\\5</A>", $line);
 240
 241     // make urls (with and without httpd) into proper links
 242     $line = eregi_replace("([[:space:]]|^)([[:alnum:]]+)://([^[:space:]]*)([[:alnum:]#?/&=])",
 243       "\\1<A HREF=\"\\2://\\3\\4\" TARGET=\"newpage\">\\2://\\3\\4</A>", $line);
 244     $line = eregi_replace("([[:space:]])www\.([^[:space:]]*)([[:alnum:]#?/&=])",
 245       "\\1<A HREF=\"http://www.\\2\\3\" TARGET=\"newpage\">www.\\2\\3</A>", $line);
 246
 247     // make email addresses into mailtos....
 248     $line = eregi_replace("([[:space:]]|^)([a-zA-Z0-9@.]+)\(([^)]+)\)",
 249       "\\1<a href=\"mailto:\\2\">\\3</a>", $line);
 250
 251     // !# at the beginning of any lines means a heading
 252     $line = eregi_replace( "^!([1-6]) (.*)$", "<h\\1>\\2</h\\1>", $line );
 253
 254     // acronym handing, example HTML(Hypertext Markyp Language)
 255     $line = ereg_replace( "([A-Z]+)\(([^)]+)\)", "<acronym title=\"\\2\">\\1</acronym>", $line );
 256
 257     // *Moodle Specific*
 258     if (IN_MOODLE==1) {
 259       // Replace resource link >>##(Description Text)
 260       $line = eregi_replace( " ([a-zA-Z]+):([0-9]+)\(([^)]+)\)",
 261          " <a href=\"".$CFG->wwwroot."/mod/\\1/view.php?id=\\2\">\\3</a> ", $line );
 262
 263       // Replace picture resource link
 264       global $course;    // This is a bit risky - it won't work everywhere
 265
 266       if ($CFG->slasharguments) {
 267         $line = eregi_replace( "/([a-zA-Z0-9./_-]+)(png|gif|jpg)\(([^)]+)\)",
 268           "<img src=\"$CFG->wwwroot/file.php/$course->id/\\1\\2\" alt=\"\\3\" />", $line );
 269       } else {
 270         $line = eregi_replace( "/([a-zA-Z0-9./_-]+)(png|gif|jpg)\(([^)]+)\)",
 271           "<img src=\"$CFG->wwwroot/file.php\?file=$course->id/\\1\\2\" alt=\"\\3\" />", $line );
 272       }
 273
 274       replace_smilies( $line );
 275
 276     }
 277
 278     return $line;
 279   }
 280
 281
 282   function spellcheck( $line,$pspell_link ) {
 283
 284     // split line into words
 285     $words = preg_split( "/[\s,-.]/ ", $line );
 286
 287     // run through words
 288     $newline = "";
 289     foreach($words as $word) {
 290       $check_word = eregi_replace( "[,;:./&()* ?\"]", "", $word );
 291       $check_word = eregi_replace( "^'|'$","",$check_word );
 292
 293       // words not to check
 294       $docheck = true;
 295       if (eregi("[0-9]",$check_word)) { $docheck=false; }
 296
 297       if ( $docheck && (!pspell_check( $pspell_link, $check_word)) ) {
 298         $suggests = pspell_suggest( $pspell_link,$check_word );
 299         $suggest_line = "";
 300         foreach($suggests as $suggest) {
 301           $suggest_line = $suggest_line . " " . $suggest;
 302         }
 303         $word = "<span class=\"spellcheck\"><acronym title=\"$suggest_line\">$word</acronym></span>";
 304       }
 305       $newline = $newline . " " . $word;
 306     }
 307
 308     return $newline;
 309   }
 310
 311
 312   function format( $content ) {
 313     // main entry point for processing TikiText
 314     // $content is string containing text with Tiki formatting
 315     // return: string containing XHTML formatting
 316
 317     // initialisation stuff
 318     $this->output = "";
 319     $this->block_state = STATE_NONE;
 320     $this->list_state = LIST_NONE;
 321     $this->list_depth = 0;
 322     $this->list_backtrack = array();
 323     $this->spelling_on = false;
 324
 325     // split content into array of single lines
 326     $lines = explode( "\n",$content );
 327     $buffer = "";
 328
 329     // add a wiki div tag for CSS
 330     $buffer = $buffer . "<div class=\"wiki\">\n";
 331
 332     // run through lines
 333     foreach( $lines as $line ) {
 334
 335       // is this a blank line?
 336       $blank_line = eregi( "^[[:blank:]\r]*$", $line );
 337       if ($blank_line) {
 338         // first end current block according to state
 339         $buffer = $buffer . $this->close_block( $this->block_state );
 340         $this->block_state = STATE_NONE;
 341         continue;
 342       }
 343
 344       // is this a spelling line
 345       $spell_parms = array();
 346       $spelling = eregi( "^!SPELL:([a-z]+):?(american|british|canadian)?(\r| |$)", $line,$spell_parms );
 347       if ($spelling) {
 348         $this->spelling_on = true;
 349         $pspell_link = pspell_new( $spell_parms[1], $spell_parms[2] );
 350         $line = "";
 351       }
 352
 353       // spellcheck
 354       if ($this->spelling_on) {
 355         $line = $this->spellcheck( $line, $pspell_link );
 356       }
 357
 358       // act now depending on current block state
 359       if ($this->block_state == STATE_NONE) {
 360         // first character of line defines block type
 361         if (eregi( "^> ",$line )) {
 362           // blockquote
 363           $buffer = $buffer . "<blockquote>\n";
 364           $buffer = $buffer . $this->line_replace( eregi_replace( "^>","",$line) ). "\n";
 365           $this->block_state = STATE_BLOCKQUOTE;
 366         }
 367         else
 368         if (eregi( "^  ",$line) ) {
 369           // preformatted text
 370           $buffer = $buffer . "<pre>\n";
 371           $buffer = $buffer . $this->line_replace($line) . "\n";
 372           $this->block_state = STATE_PREFORM;
 373         }
 374         else
 375         if (eregi("^\% ",$line) ) {
 376                 // preformatted text - no processing
 377                 $buffer = $buffer . "<pre>\n";
 378                 $buffer = $buffer . eregi_replace( "^\%","",$line) . "\n";
 379                 $this->block_state = STATE_NOTIKI;
 380         }
 381         else
 382         if (eregi("^Q\. ",$line) ) {
 383           // Question - para with a question class
 384           $buffer = $buffer . "<p class=\"question\">\n";
 385           $buffer = $buffer . eregi_replace( "^Q. ","",$line) . "\n";
 386           $this->block_state = STATE_PARAGRAPH;
 387         }
 388         else
 389         if (eregi("^A\. ",$line) ) {
 390           // Answer - para with an answer class
 391           $buffer = $buffer . "<p class=\"answer\">\n";
 392           $buffer = $buffer . eregi_replace( "^A. ","",$line ) . "\n";
 393           $this->block_state = STATE_PARAGRAPH;
 394         }
 395         else {
 396           // ordinary paragraph
 397           $buffer = $buffer . "<p>\n";
 398           $buffer = $buffer . $this->line_replace($line) . "\n";
 399           $this->block_state = STATE_PARAGRAPH;
 400         }
 401         continue;
 402       }
 403
 404       if (($this->block_state == STATE_PARAGRAPH) |
 405           ($this->block_state == STATE_BLOCKQUOTE) |
 406           ($this->block_state == STATE_PREFORM) ) {
 407         $buffer = $buffer . $this->line_replace($line) . "\n";
 408         continue;
 409       }
 410       elseif ($this->block_state == STATE_NOTIKI) {
 411         $buffer = $buffer . $line . "\n";
 412       }
 413     }
 414
 415     // close off any block level tags
 416     $buffer = $buffer . $this->close_block( $this->block_state );
 417
 418     // close off wiki div
 419     $buffer = $buffer . "</div>\n";
 420
 421     //return $buffer;
 422     return $buffer;
 423   }
 424
 425 }
 426
 427 ?>